hashavshevet-data-tools
Pass
Audited by Gen Agent Trust Hub on Apr 14, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes untrusted external data files (e.g., Hashavshevet .dat files, bank CSVs), which creates a potential surface for indirect prompt injection attacks where data content could influence the agent's behavior.
- Ingestion points: Data enters the agent context through the reading of local accounting files (PKUDOT.dat, HESHIN.dat, etc.) and CSV/Excel exports using Python scripts.
- Boundary markers: The provided code and instructions lack explicit delimiters or instructions to the agent to disregard potential instructions embedded within the accounting records.
- Capability inventory: The skill has access to file system operations (Read, Write, Edit) and Python execution via Bash, which could be leveraged if an injection is successful.
- Sanitization: While the code performs structural parsing (fixed-width slicing and CSV mapping), it does not include security-specific sanitization to filter out prompt injection payloads in text fields.
Audit Metadata